Use case · Dental practice
Your website, your Google listing and your booking portal describe the same practice with different hours and treatments. When a patient asks an AI assistant for a dentist, the answer draws on them: here’s how to see the difference and fix it.
Before the correction
A patient asks for a dentist for an implant, or one open on Saturday. The assistant doesn’t visit the practice: it reads the website, the map listing, the booking portal and the reviews, then composes an answer.
The questions that reach assistants are concrete and local, like the family practice open on Saturday in a given city. The trouble starts when the sources don’t say the same thing: the website shows the summer hours, the listing the winter ones, the portal a treatment you no longer offer. The assistant doesn’t know which is right, picks one and presents it as a fact.
A demo dental practice compared topic by topic: what the website, listing and portal said before alignment and what they say after.
| Before | After | |
|---|---|---|
| Saturday hours | The website says closed, the listing open in the morning, the portal doesn’t show them. | Open in the morning, written the same way on all three sources. |
| Treatments | The portal still lists a treatment suspended months ago. | A single list, with the date of the last update. |
| Recent reviews | The most recent dates back many months. | Regular invitations after the visit, to those who agreed to be contacted again, the same for everyone and with nothing in return. |
The practice’s measurement
The platform repeats your patients’ local questions and compares what the assistant states with what you publish. Each difference becomes a specific task, with the point to correct and the text ready.
AI visibility records at every cycle whether the practice is named, in what position and with which sources. AI claims isolates the facts, such as opening hours, treatments and insurance agreements, and places them next to the source they come from. Review monitoring reads the location’s listing: dates, ratings, replies given and not given. There is no need to guess which listing to correct first: the answer itself tells you.
The task created when an assistant’s answer reports opening hours different from those published by the practice: the wrong fact, the right one, where to correct and when to check.
After the correction
First come the facts a patient uses to decide, namely hours and treatments; then the reviews, which show whether the practice is active. Afterwards, the check compares two periods and shows which actions touched your questions.
The glimpse shows the Actions check on a demo brand: for each action what is measured, whether the approved text is actually published and whether the same measure changes at the next collection, on the same questions. The two stay separate, because one does not prove the other, and a change is presented as a reading, not as proof of an effect.
Avoid: no reviews for months, and the practice no longer appears among those suggested in the city. Do: where the rules of your professional body allow it, an invitation to review after the visit to patients who agreed to be contacted again, the same for everyone and without incentives, with a reply from the practice to every comment, with no health data.
Information, not advertising
For a dental practice, the correction remains health information: treatments, qualifications, opening hours, how to access care. Health information has its own rules in every country, and a text written only to get noticed by assistants should be reread in light of those rules.
In Italy the rules on health information are set out in the law and in the guidance of the professional body; in Germany in the law on advertising in the healthcare field. The practical criterion is the same one that helps assistants: a page that describes what is done, who it is suitable for and who performs it is also the clearest text to cite.
A sample answer to a question about a treatment, with three parts flagged: the practice named, the line taken from the website that should be rewritten as information and the page cited as the source.
QuestionA dentist for teeth whitening near me?
AI assistant · sample answer
You could try the dental practice on Via dei Tigli, whose website talks about a new smile in a single session. The information comes from the practice’s treatments page.
The demo practice makes it into the answer: that is the presence signal.
The line repeated is a promise, not information: it should be rewritten to describe the treatment.
The source is the practice’s website: correcting that page changes what the assistant picks up.
Frequently asked questions
Why another practice gets recommended, whether you need different pages for each assistant, how to ask for reviews and what stays private.
Possible causes include inconsistent sources, few recent reviews and a website that doesn’t say in one sentence what you do and where. The measurement on your panel shows which of these weighs in your case.
No. Assistants read the same public sources: one consistent version of the facts, identical on the website, the listing and the portal, serves them all.
Yes, if the rules of your professional body allow it: to everyone in the same way, without giving anything in return and only to patients who agreed to be contacted again. Replies to reviews never mention a patient’s health data, nor confirm that the person is your patient.
No. Your practice’s measurement stays in your private area, and no practice is ever named on public pages.
There is no fixed time: it depends on when the sources are reread. That is why the same question is repeated at every cycle and the platform shows you the difference, when there is one.
Related paths
Each use case starts from a specific business. From here you move on to the other businesses, to the pages of the sectors they belong to and to the method we use to measure answers.
Use cases by type of business: the questions people ask AI assistants, what TrustInsider measures and what you can then fix.
ReadHow TrustInsider works in six steps: brand, questions, quote, evidence, page-by-page intervention and a check on the same questions.
ReadHow every TrustInsider measure is built: sampling of AI answers, sources and rights, index calculation and declared limits.
ReadHelp with access, credits, collections, integrations and data: guides, FAQs, the site assistant and support requests from your workspace.
Read